Monitoring Morphology and Currents at the Hatteras Breach

Transcription

1 Monitoring Morphology and Currents at the Hatteras Breach By Ty V. Wamsley U.S. Army Engineering Research and Development Center Coastal and Hydraulics Laboratory 3909 Halls Ferry Road, Vicksburg, MS Ty. V. usace.army.m il Kent K. Hathaway U.S. Army Engineering Research and Development Center Coastal and Hydraulics Laboratory Field Research Facility, Duck, NC ABSTRACT On 18 September 2003, Hatteras Island, North Carolina, was breached by Hurricane Isabel about 6 miles northeast of Hatteras Inlet to connect Pamlico Sound with the Atlantic Ocean. The breach occurred at a location of minimum island width and elevation. As part of research on coastal inlets, the new opening was surveyed to provide much-needed data on the hydrodynamic and morphologic evolution of barrier island breaches. Two topographic and shallow water surveys were conducted I 0 days apart to capture short-term temporal changes in the breach morphology. Analysis of the survey data indicates rapid morphology change. Ebb and flood shoals formed within 2 weeks of the breach. The main breach channel widened by as much as 125 ft and migrated to the west by as much as 80 ft in the 10 days between surveys. Water levels on both the ocean and sound side of the barrier were also measured. Current velocities were measured with surface drifters and with an ADCP during two field deployments. Maximum current velocities were on the order of7 ftlsec. The surveys provide quantitative data on the evolution of barrier island breaches that will be applied in development of numerical models of coastal breaching. Additional Keywords: ADCP, barrier island, bathymetry, Hurricane Isabel, inlet, Pamlico Sound, survey INTRODUCTION H urricane Isabel made landfall along the North Carolina Outer Banks on 18 September The eye came ashore between Cape Lookout and Ocracoke Island, near Drum Inlet. Isabel was a Category 2 storm on the Saffir-Simpson scale with maximum winds of approximately 100 mph and produced storm surges 6.5 to 8 ft above normal tide level near the point of landfall. The storm breached Hatteras Island about 6 miles northeast of Hatteras Inlet between the villages of Hatteras and Frisco (Figure 1). The Hatteras breach quickly widened to an overall width of over I,500 ft. The breach contained two "breach islands" that created three distinct breach channels (Figure 2). During a 1933 storm, a breach occurred at nearly the same location. A bridge was being built over the new inlet in the 1930s, but the inlet closed naturally before the bridge was finished and construction was stopped. The U.S. Army Corps of Engineers (USACE), Wilmington District mechanically closed the breach created by Hurricane Isabel on I November 2003 (Wutkowski 2004). This section of Hatteras Island is characterized by having medium-fine sand from the mean high water (MHW) line to 6 ft depths. The average significant wave height, as determined by the Wave Information Study (WTS) Level 3 for I 0 years ( ) at station 262, is 3.6 ft with a standard deviation of 2.3 ft. Mean wave period is 6 sec with a standard deviation of 2.4 sec, Figure 1. Location map for Hatteras breach. and the predominant wave direction is from south to southeast. The mean tide range in this area, measured at the Cape Hatteras Fishing Pier, is 3.05 ft and spring tide range is 3.53 ft. A coastal breach is a new opening in a narrow landmass such as a barrier island that allows water to flow between water bodies on either side of the barrier. Unintended breaches occur around the coast of the United States each year and are a serious concern in developed areas or areas of critical habitat (Kraus and Figure 2. Hatteras breach, 22 September 2003 (USACE Wilmington District photograph). Shore & Beach Vol. 72, No. 2, Spring 2004, pp

2 Report Documentation Page Form Approved OMBNo ~lie ~g bunlen for the collection of infonnation is estimated to average I how" per response, including the time for reviewing instructions, searching existing data sources, gathering and IIUliJI1aUUDg the data needed, and completing and reviewing the collection of infonnation. Send comments reganling this burden estimate or any other aspect of this collection of infonnation, including suggestions for redw:ing this burden, to Washington Headquarters Services, Directorate for Information Opemtions and Reports, 1215 Jefferson Davis Highway, Suite 1204, Arlington VA Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to a penalty for failing to comply with a collection of infonnation if it does not display a CW"rently valid OMB control number. I. REPORT DATE 2. REPORT TYPE 3. DATES COVERED 2004 N/A - 4. TITLE AND SUBTITLE Sa. CONTRACT NUMBER Monitoring Morphology and Currents at the Hatteras Breach Sb. GRANT NUMBER Sc. PROGRAM ELEMENT NUMBER 6. AUTHOR(S) Sd. PROJECT NUMBER Wamsley, Ty V., and Hathaway, Kent K. Se. TASK NUMBER Sf. WORK UNIT NUMBER 7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) 8. PERFORMING ORGANIZATION Coastal and Hydraulics Laboratory, US Army Engineering Research and REPORT NUMBER Development Center, Vicksburg, MS; Field Research Facility, Coastal and Hydraulics Laboratory, US Army Engineering and Research Center, Duck,NC. 9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSOR/MONITOR'S ACRONYM(S) 12. DISTRIBUTION/AVAILABILITY STATEMENT Approved for public release, distribution unlimited 13. SUPPLEMENTARY NOTES 14. ABSTRACT I I. SPONSOR/MONITOR'S REPORT NUMBER(S) On 18 September 2003, Hatteras Island, North Carolina, was breached by Hurricane Isabel about 6 miles northeast of Hatteras Inlet to connect Pamlico Sound with the Atlantic Ocean. The breach occurred at a location of minimum island width and elevation. As part of research on coastal inlets, the new opening was surveyed to provide much-needed data on the hydrodynamic and morphologic evolution of barrier island breaches. Two topographic and shallow water surveys were conducted 10 days apart to capture short-term temporal changes in the breach morphology. Analysis of the survey data indicates rapid morphology change. Ebb and flood shoals formed within 2 weeks of the breach. The main breach channel widened by as much as 125ft and migrated to the west by as much as 80ft in the 10 days between surveys. Water levels on both the ocean and sound side of the barrier were also measured. Current velocities were measured with surface drifters and with an ADCP during two field deployments. Maximum current velocities were on the order of 7 ft/sec. The surveys provide quantitative data on the evolution of barrier island breaches tha.t will be applied in development of numerical models of coastal breaching. 1S. SUBJECT TERMS 16. SECURITY CLASSIFICATION OF: 17. LIMITATION OF IS. NUMBER ABSTRACT OF PAGES a. REPORT b. ABSTRACT c. TillS PAGE uu 6 unclassified unclassified unclassified " 19a.NAMEOF RESPONSIBLE PERSON Standard Form 298 (Rev. 8-98) l'r<saibedbyansistdz39-18

3 Wamsley 2003). The Hatteras breach destroyed utility infrastructure and severed North Carolina Highway 12, isolating Hatteras Village from the rest of Hatteras Island. Parking lots and buildings near the breach were also destroyed. Highway 12 is the only transportation route east from the village. Residents were unable to drive to work and home. Businesses at Hatteras Village and on Ocracoke Island experienced loss of tourism revenue because of the restricted access. Despite the potential societal and environmental costs associated with breaches, relatively little information is known on the physical processes of breaches. The Hatteras breach was in close proximity to the USACE Research and Development Center, Field Research Facility (FRF), which provided an opportunity to study large breaches of barrier islands. Highresolution surveys of the breach produced needed data on the hydrodynamic and morphologic evolution of barrier breaches. The surveys not only qualitatively describe the course of breach evolution, but also provide quantitative data for numerical models of coastal breaching under development by the USACE. Two topographic and shallow water surveys were conducted to capture the short-term temporal changes in breach morphology. The first survey was conducted over 3-5 October A single multi-beam survey was conducted on 5 October. The second survey was conducted from October. The bathymetry surveys of the breach channels, margin shoals, and flood shoal were performed with a Real-Time Kinematic (RTK) GPS Waverunner survey system. The system is designed for shallow water applications and can reach areas not accessible with more conventional survey methods. Topography was surveyed with an RTK-GPS system to clearly identify the shoreline of the beach and breach edge. The "breach islands" in the middle of the breach were also topographically surveyed to capture their slopes and shorelines. The multi-beam survey was performed to measure the size and extent of the ebb shoal. The multi-beam system also surveyed the main breach channel. Freeman, Bernstein, and Mitasova (2004) give a complete discussion of the survey plan and techniques. During the first survey period, drogues were timed to approximate surface currents through the main breach channel. Drogue measurements were made at various times throughout the tidal cycle over 4-6 October The current was also measured with an Acoustic Doppler Current Profiler (ADCP) in the breach during two field deployments. The first current survey took place over October near the time the second morphology survey was being conducted. The current was surveyed again on 24 October. Water levels were also recorded near the breach on both the Atlantic Ocean side and the Pamlico Sound side from 3 October to 12 November The purpose of this paper is to document the short-term morphologic evolution and hydrodynamic characteristics of the breach based on an analysis of the collected data. OVERVIEW OF BREACHING PROCESS Breaching potential is maximized if the barrier is low and narrow. Narrowing of the barrier results from shore erosion on either the ocean or bay/sound side. Lowering of the barrier results from dune degradation. Several causes of dune degradation can be identified, including fixed footpaths for beach access, seepage, undercutting and failure from wave attack, and wave overtopping. The narrowing and lowering of the barrier creates localized low profiles in the dune system. When water levels are elevated, inundation occurs and water begins to flow through the localized low areas. Once the dune crest is submerged, erosion occurs rapidly and can be catastrophic. After the complete wash out of the dune, a breach widens by erosion and collapse of the bank and deepens as flow scours the channel. Margin shoals are often formed immediately after breach opening. If the breach scour is sufficiently deep, water flow can occur between the two water bodies on each side of the barrier, even after the storm subsides. Tidal flow then continues to widen and deepen the breach channel. If the breach flow is strong enough to flush littoral drift-derived sediments from the breach faster than it is introduced, the breach is maintained and a tidal inlet formed. The flushing of sediment by the tidal current may create an ebb and/or flood shoal. Hatteras Island was breached by elevated water levels and wave attack from the ocean side. The barrier is approximately 500 ft wide where the breach occurred, one of the narrowest sections along Hatteras Island. The long-term shoreline recession rate at this location is about 3 to 4 ftlyear (Overton and Fisher 2004). Airborne LIDAR surveys by the U.S. Geological Survey (USGS) and National Aeronautics and Space Administration (NASA) show that the breach occurred not only at a location of minimum island width, but also at a minimum of island elevation (Sallenger 2004 ). The primary cause of the low dune elevations is likely wave attack as a result of the receding shoreline. However, beach access may also have contributed. The breach occurred near a parking lot for beach goers. Dunes at access points are often lower due to foot traffic. If possible, coastal mangers should discourage beach access at narrow regions of barrier islands and establish access points where the barrier island is wide. The narrowing and lowering of Hatteras Island at this location weakened the barrier and subjected it to breaching. Evidence of powerful overtopping water flow was observed at the breach. Brush and other vegetation adjacent to the breach channels were matted and flattened toward the Pamlico Sound by overwashing flow. Channelization at the lowest points in the barrier scoured the breach channels. BREACH MORPHOLOGY AND SHORT-TERM EVOLUTION The morphology and short-term evolution of the breach were examined through spatial surface analysis of three-dimensional digital elevation models (DEMs) created with high-resolution topographic and bathymetric data. Depths over the ebb shoal were collected on 5 October, in addition to the surveys of 3-5 October (Survey I) and October (Survey 2). Survey 2 had l.o... 4.G ICAV081) Pamllco Sound FloodShoill Figure 3. Digital elevation model of Hatteras Breach. 10 Shore & Beach Vol. 72, No. 2, Spring 2004, pp. 9-14

4 depths in the west channel were 7 to I 0 ft NAVD88. The eastern most breach channel was approximately 325 ft wide on the sound side and 350 ft wide on the ocean side and was the main channel as it captured a majority of the tidal prism. The unrestricted flow through this channel created scour depths down to 20 ft NAVD88. Figure 4. Western edge of peat terrace, looking southeast. The peat extends approximately 1.5 to 2ft above the adjacent sandy bottom. The flood shoal is readily identifiable in Figure 3. The Survey DEM showed that a flood shoal formed within 2 weeks after the breach opened with the centroid approximately I, 750 ft from Highway 12. The shallowest water over the flood shoal was less than 0.5 ft NAVD88 and had to be surveyed by wading, as depths were too small even for the waverunner system. The second survey showed little growth of the flood shoal. A volumetric analysis indicates that the flood shoal gained on the order of I 0,000 cu yd of sand over the I 0-day period. A well-defined ebb shoal also formed by 5 October (Figure 3). The ebb shoal extended offshore as far as I,250 ft from the former location of the highway. Depths over the ebb shoal were 4 to 6ft NAVD88. Waves were often observed breaking on the ebb shoal. The DEM analysis indicates rapid morphology change. Figure 6 is a comparison plot of breach cross-sections from Surveys I and 2. Comparisons are made only where sufficient coverage of the channel was captured by both surveys. Due to debris and wave conditions, the seaward end of the west breach channel was not measured in Survey I. As Figure 6 shows, there was little change in the west channel, because of the substantial amount of armoring by debris and peat. Flow velocities were significantly weaker than observed in the main channel. The sound end of the Figure 5. West breach channel, looking east, 5 October Note the old bridge pilings on the right and the extensive road debris in the channel. best coverage ofthe nearshore region just seaward of the breach openings and, combined with the ebb shoal data, provides the most comprehensive data set. Figure 3 is a DEM created for illustrative purposes from Survey 2 data combined with the multi-beam ebb shoal data. Three breach channels, separated by breach islands, are well defined. The breach islands were not washed away because of resistance of peat outcroppings that fronted the islands on the ocean side (Figure 4). The peat acted as an erosion-resistant barrier, revetting the island at this location. The broad peat terrace extended ftom the eastern most breach island, across the middle breach channel and west breach island (Figure 3). The peat terrace restricted flow in the middle breach channel, which was about 225 ft wide. Water flowed through the middle channel and over the peat terrace only at higher tide elevations. The restricted flow resulted in little scour of this channel. Maximum depths in the middle channel were about 5 ft NAVD88 ( 1988 North American Vertical Datum). The west channel was about 350 ft wide and littered with debris (Figure 5). Pilings from the bridge partially constructed in the 1930s extended across the channel on the ocean side. Large chunks of asphalt and roadbed material were visible at lower tide elevations, particularly on the west side of the channel. Near the center of the channel was a peat outcropping with large chunks of asphalt resting on top. The flow through the west channel was somewhat greater than that of the middle channel. Survey 2 maximum Transect Survey Survey 2-20~--~--L---~--~--~--~--~--~ BOO ,_..,..., Figure 6. Comparison of Survey 1 and Survey 2 breach channel cross-sections (North Carolina Department of Transportation (NCDOT) photograph). Shore & Beach Vol. 72, No.2, Spring 2004, pp

5 .J ~ IT\ :11 ~1) 1: h 1>.:: 1 I 1C<I ' ti 'J! ~"."l Figure 7. Comparison or the 0-contour INAVD88). The aerial photograph was taken on 24 September, tide elevation -1.3 It NAVDBBINCDOT photograph). ~ rt ll=' ". l l 'ln Figure 8, Pamllco Sound and ocean water level dltrerence, J October to 12 November middle channel shoaled approximately 2.5 ft in the I 0 days between surveys. The sediment li ke ly originated at the margin shoals on the sound side and could not be flushed out by flow restricted by the peat terrace. The central and seaward portions of the middle channel did not change between Surveys I and 2. The main breach channel experienced the greatest amount of change. Defining breach width as the distance across the channel between the NAVD88 0-contour (which corresponds to ft msl), the breach widened by approximately 125 ft on the ocean side, 65ft through the middle part of the channel, and 25ft on the sound side. From the center of the barrier toward the sound, the main channel migrated approximately 20 ft to the west over the I 0-day period whi le on the ocean side the channel migrated over 80 ft to the west. The average channel depths were maintained and maximum channel depths generally increased through the middle part of the channel and decreased on the sound side over the I 0-day period. The average cross-sectional area of the breach channel increased from 3,400 to 4,000 sq ft. northerly direction and also elongated by approximately 65 ft. The 0-contour on the front side of each island receded. This recession, however, was not from erosion of the peat terrace. The peat terrace elevation was below 0 NAVD88, and the sandy material on top eroded during higher tide elevations. Figure 7 also shows beach shoreline change within about 500 ft of the breach. On the west side, the beach eroded approximately 35 ft. Sediment supplied by longshore transport Figure 7 is a comparison of the 0-contour (NAVD88) between Surveys I and 2. The recession of both channel banks at the ocean end of the breach is evident. The east bank receded 25 to 45 ft while the west bank moved 40 to 80 ft. On the sound end of the breach channel, the east bank advanced approximately 25 ft, pushing the channel to the west. This end of the channel also continued to widen, however, as the west bank receded about 45 ft. The prograding shoulder of the barrier island was supplied sediment from the ocean end of the breach bank and the sound side of the barrier island. A strong current was observed in the margin channel on the east side of the breach. Bank erosion was easily observed at this location. As the tide fell, a scarp formed on the sound side of the barrier. During lower tide elevations, the strong flow removed sand from the base of the scarp until a tension crack formed, leading to mass fa ilure. The sandy bank material slumped into the margin channel, moving down the bank slope where the tidal current carried it downstream. At higher tide elevations, the scarp became submerged and was smoothed by the current flow and wave action. Recession of the 0-contour as a result of this process is seen in Figure 7. The maximum shoreline recession measured over the I 0-day period at the margin channel was approximately 20 ft. Figure 7 illustrates the elongation of the margin shoals behind both breach islands. The east island shoal advanced approximately 65 ft. The west island shoal migrated in a 12 Shore & Beach Vol. 72, No. 2, Spring 2004, pp.